Pediatric gastrointestinal (GI) bleeding is a relatively common problem which carries a substantial risk of morbidity and mortality. A small series from Montreal demonstrated a 1.6% rate of gastrointestinal bleeding in ICU patients, but only limited characterizations of endoscopic interventions or other management strategies have been described in non-variceal GI bleeding [12]. Upper GI tract bleeding has been described in numerous conditions, commonly due to erosions, ulcers or vascular malformations (Figs. 11.1, 11.2, 11.3 and 11.4). However, across pediatric age groups, the etiologies of GI bleeding may differ (Table 11.4). It is also important to note that it is not uncommon to have lower gastrointestinal bleeding in children due to an upper GI source.

Current management of pediatric gastrointestinal bleeding employs a variety of medical, transfusion and endoscopic strategies. Red blood cell transfusion is also often necessary in the setting of acute blood loss caused by GI bleeding and should be available in the operating room. Additionally, fresh frozen plasma, platelets and other agents may be required. The major limitations in managing these cases are the size of the tools used for bleeding control. The majority of equipment including hemostatic clips and multipolar coagulation probes do not fit through the working channel. Also, the typical variceal band ligation cap (typically used in adults) is difficult to pass in children under 10 kg in size. Commonly used strategies for bleeding control in smaller patients may include an injection needle that can pass through a 2.0-mm channel or the argon plasma probe which can also pass through a smaller channel (Video 11.1). Alternatively, an attempt at using a larger caliber endoscope can be considered depending on the anticipated treatment. In our experience, we typically make a diagnostic assessment with the smallest appropriate endoscope and change endoscopes if needed for an intervention.

Celiac disease (CD) is an autoimmune condition involving the inability of the small intestine to properly respond to the breakdown and absorption of gluten. In celiac disease, the host’s immune system is activated with the ingestion of gluten causing local inflammation, damage and destruction of the small intestine. Over time, this can lead to classic symptoms of celiac disease including failure to thrive, weight loss, diarrhea, abdominal pain and malnutrition.

This disease is particularly important to pediatric gastroenterologists because the classic symptoms typically begin during the childhood and adolescence. Studies performed in the USA and Europe estimate the prevalence to be 3–13/1000 or 1:300–1:80 of children between 2.5 and 15 years of age [13]. This understanding of the epidemiology of CD has greatly changed over the past few decades due to serologic tests in addition to advances in endoscopy. Until recently, it was not appreciated that a large majority of individuals can have a more chronic and insidious onset with milder symptoms than the classic, historical presentation thus making the diagnosis much more difficult [14].

Along with the changing epidemiology of CD, the diagnosis has also been a topic of numerous changes specific to endoscopy. Prior to the 1960s, the diagnosis of celiac disease was based solely on symptoms. In 1969, the European Society for the Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) developed the first diagnostic criteria for celiac disease. This included: (1) “structurally abnormal jejunal mucosa when taking a diet containing gluten,” (2) “clear improvement of villous structure when taking a gluten-free diet,” (3) “deterioration of the mucosa during challenge.” With these criteria there was a minimum of three sets of biopsies required over a period of at least one year. The initial biopsy was performed to demonstrate damage, the second biopsy was performed after one year on a gluten-free diet to document healing and the third biopsy was performed after a few months of gluten re-introduction. Further revisions in 1990 stated that demonstrating the recurrence of histologic abnormalities after a gluten-free diet and gluten challenge was unnecessary as a part of the diagnostic criteria [15].

Histopathology remains the mainstay of the diagnosis for celiac disease. A clinical diagnosis alone was shown to be incorrect in >50% of cases [13]. In addition, celiac disease is a chronic condition involving a lifelong abstinence from gluten-containing products. Therefore, it is highly recommended that the diagnosis be confirmed with endoscopy and histologic evaluation before the treatment plan is initiated. Although symptoms may be “classic” there are other conditions that may manifest with similar symptoms. A recent report from ESPGHAN proposes a possibility of deferring biopsy if the serology is abnormal, and the patient has a specific HLA haplotype [16]. This was evaluated in a Canadian cohort that challenges this concept as an alternative diagnosis may be missed if strictly adhering to the proposed European guidelines [17]. Thus, the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) continues to recommend upper endoscopy with intestinal biopsy for confirmation of celiac disease.

Celiac disease is patchy in nature with differences in severity at different locations in the small bowel [18]. For this reason, multiple duodenal biopsies need to be obtained in order to increase the probability of accurately confirming a diagnosis. Important locations to biopsy from include: the duodenal bulb, as well as distal segments of the duodenum, and duodenal–jejunal junction. At least six duodenal biopsies including at least two bulb biopsies should be performed to appropriately evaluate for celiac disease. This may be even more important in patients with early disease or milder symptoms, as their small intestinal architecture may not be evident endoscopically (Video 11.1).

Endoscopic features of CD can be varied. In a patient with mild disease, the mucosa may appear to be completely normal. However, in patients with severe disease, it is not uncommon to visualize diffuse disturbances in duodenal architecture. The most common macroscopic findings of duodenal villous atrophy include the absence of mucosal folds (Fig. 11.5a), scalloping (Fig. 11.5b) and submucosal blood vessels and a mosaic pattern of the mucosa [13]. Although these are most common, it is again important to note that affected patients may need subtotal to total villous atrophy for these to become visible endoscopically. In addition, the phrase “all that scallops is not celiac disease” has been coined to illustrate that celiac disease is only one of the disease states implicated with these gross abnormalities.

An experienced pediatric pathologist with specific experience with the evaluation of gastrointestinal mucosa should perform the evaluation of any obtained biopsy specimens; this increases the likelihood of an accurate diagnosis. The modified Marsh criteria (Table 11.5) are widely used for classifying histologic changes. It reveals four categories: pre‐infiltrative (type 0), infiltrative (type 1), infiltrative‐hyperplastic (type 2) and flat‐destructive (type 3) and the atrophic‐hypoplastic (type 4) lesion [19]. These criteria establish a grading severity for mucosal destruction and are not specific to celiac disease. However, with the appropriate clinical correlate, a diagnosis of celiac disease can be made. Characteristically there are increases in intraepithelial lymphocytes (>30 lymphocytes/100 enterocytes), decrease in goblet cells, brush border abnormalities and villous flattening.

Eosinophilic esophagitis (EE) is a disorder isolated to the esophagus associated with eosinophilic infiltration in association with upper gastrointestinal symptoms. It is a relatively newly described disorder, and the incidence has increased over the past 20 years. As studied by Noel et al. [20] in the early 2000s, there was a fourfold increase in pediatric EE prevalence in the Midwest United States with a relative incidence of ~1/10,000 children per year. It is not uncommon for a patient to be initially (and incorrectly) diagnosed with gastroesophageal reflux (GERD) but then found not to respond to typical anti-reflux medications including proton pump inhibitors and H2 receptor antagonists [21].

EE can present in either childhood or adulthood; however, there are generally differences in the presentations [22]. Adults commonly present with intermittent dysphagia (29–100%) and food impaction (25–100%) [20]. Children may have nonspecific symptoms due to the inability to describe the sensation in the esophagus. Mostly commonly, the presenting symptoms are similar to GERD including heartburn or regurgitation. Other presenting symptoms in children that differ from adults include emesis, abdominal pain, failure to thrive, and diarrhea. Food impaction may occur more frequently as a presenting finding in adults, but this is not uncommon in the pediatric population. The pathologic process is unknown; however, it seems to be separate from GERD given that it does not lead to mucosal destruction or ulceration even in severe cases. Unlike GERD, which when progressive can lead to esophageal adenocarcinoma, there does not seem to be a link between EE and carcinoma.

In the past, it was common that in the absence of endoscopic findings of esophagitis only distal esophageal biopsies were obtained to evaluate EE (if at all). Newer recommendations advise the utility of biopsies in additional areas of the esophagus (mid and proximal) [22]. From these recent studies, it has been found that histopathologic abnormalities are common in biopsy specimens of even normal appearing mucosa. For this reason, it is also recommended that biopsies be taken in multiple levels of the esophagus even if it is grossly normal. In addition, biopsies should also be obtained from the stomach and duodenum to assess for other disease entities such as eosinophilic gastroenteritis and inflammatory bowel disease.

Eosinophilic esophagitis, as with many other GI diseases, can have completely normal mucosa grossly. This does not exclude the diagnosis of EE. When there is more severe disease, there are common features seen on endoscopy. These include vertical lines, linear furrowing, white exudates, circular rings aka “felinization” or “trachealization,” and strictures (Figs. 11.6 and 11.7). These are not pathognomonic of EE but can be helpful in diagnosis in the correct clinical context.